lm832x: Take DeviceState pointer in lm832x_key_event()

Since lm832x has been qdev'ified, its users will generally
have a DeviceState pointer rather than an i2c_slave pointer,
so adjust lm832x_key_event's prototype to suit.

This allows the n810 (its only user) to actually pass a correct
pointer to it rather than NULL. The effect is that we no longer
segfault when a key is pressed.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
NB: this patch depends on the OMAP GPIO v2 patchset.

 hw/i2c.h     |    2 +-
 hw/lm832x.c  |    4 ++--
 hw/nseries.c |    7 +++----
 3 files changed, 6 insertions(+), 7 deletions(-)

On 07/28/2011 09:47 AM, Peter Maydell wrote:
> Since lm832x has been qdev'ified, its users will generally
> have a DeviceState pointer rather than an i2c_slave pointer,
> so adjust lm832x_key_event's prototype to suit.
>
> This allows the n810 (its only user) to actually pass a correct
> pointer to it rather than NULL. The effect is that we no longer
> segfault when a key is pressed.
>
> Signed-off-by: Peter Maydell<peter.maydell@linaro.org>
> ---
> NB: this patch depends on the OMAP GPIO v2 patchset.

Could you put together a pull request for these devices?

Regards,

Anthony Liguori

>
>   hw/i2c.h     |    2 +-
>   hw/lm832x.c  |    4 ++--
>   hw/nseries.c |    7 +++----
>   3 files changed, 6 insertions(+), 7 deletions(-)
>
> diff --git a/hw/i2c.h b/hw/i2c.h
> index 5514402..9381d01 100644
> --- a/hw/i2c.h
> +++ b/hw/i2c.h
> @@ -72,6 +72,6 @@ void wm8750_set_bclk_in(void *opaque, int new_hz);
>   void tmp105_set(i2c_slave *i2c, int temp);
>
>   /* lm832x.c */
> -void lm832x_key_event(i2c_slave *i2c, int key, int state);
> +void lm832x_key_event(DeviceState *dev, int key, int state);
>
>   #endif
> diff --git a/hw/lm832x.c b/hw/lm832x.c
> index 590a4cc..992ce49 100644
> --- a/hw/lm832x.c
> +++ b/hw/lm832x.c
> @@ -474,9 +474,9 @@ static int lm8323_init(i2c_slave *i2c)
>       return 0;
>   }
>
> -void lm832x_key_event(struct i2c_slave *i2c, int key, int state)
> +void lm832x_key_event(DeviceState *dev, int key, int state)
>   {
> -    LM823KbdState *s = (LM823KbdState *) i2c;
> +    LM823KbdState *s = FROM_I2C_SLAVE(LM823KbdState, I2C_SLAVE_FROM_QDEV(dev));
>
>       if ((s->status&  INT_ERROR)&&  (s->error&  ERR_FIFOOVR))
>           return;
> diff --git a/hw/nseries.c b/hw/nseries.c
> index 32f2f53..45b52bb 100644
> --- a/hw/nseries.c
> +++ b/hw/nseries.c
> @@ -45,7 +45,7 @@ struct n800_s {
>       i2c_bus *i2c;
>
>       int keymap[0x80];
> -    i2c_slave *kbd;
> +    DeviceState *kbd;
>
>       TUSBState *usb;
>       void *retu;
> @@ -362,7 +362,6 @@ static int n810_keys[0x80] = {
>   static void n810_kbd_setup(struct n800_s *s)
>   {
>       qemu_irq kbd_irq = qdev_get_gpio_in(s->cpu->gpio, N810_KEYBOARD_GPIO);
> -    DeviceState *dev;
>       int i;
>
>       for (i = 0; i<  0x80; i ++)
> @@ -375,8 +374,8 @@ static void n810_kbd_setup(struct n800_s *s)
>
>       /* Attach the LM8322 keyboard to the I2C bus,
>        * should happen in n8x0_i2c_setup and s->kbd be initialised here.  */
> -    dev = i2c_create_slave(s->i2c, "lm8323", N810_LM8323_ADDR);
> -    qdev_connect_gpio_out(dev, 0, kbd_irq);
> +    s->kbd = i2c_create_slave(s->i2c, "lm8323", N810_LM8323_ADDR);
> +    qdev_connect_gpio_out(s->kbd, 0, kbd_irq);
>   }
>
>   /* LCD MIPI DBI-C controller (URAL) */

On 29 July 2011 15:31, Anthony Liguori <anthony@codemonkey.ws> wrote:
> On 07/28/2011 09:47 AM, Peter Maydell wrote:
>>
>> Since lm832x has been qdev'ified, its users will generally
>> have a DeviceState pointer rather than an i2c_slave pointer,
>> so adjust lm832x_key_event's prototype to suit.
>>
>> This allows the n810 (its only user) to actually pass a correct
>> pointer to it rather than NULL. The effect is that we no longer
>> segfault when a key is pressed.
>>
>> Signed-off-by: Peter Maydell<peter.maydell@linaro.org>
>> ---
>> NB: this patch depends on the OMAP GPIO v2 patchset.
>
> Could you put together a pull request for these devices?

Sure. Do you want just the GPIO patches and this patch,
or shall I put the nand/onenand patches in too?

thanks
-- PMM